Download House M.D. Transcripts

download.rb

#!/usr/bin/env ruby

require 'rubygems'
require 'nokogiri'
require 'open-uri'

INDEX_PAGE = "https://clinic-duty.livejournal.com/12225.html"

# Use URI.open for URLs
(URI.open(INDEX_PAGE) { |f| Nokogiri::HTML(f) }/"table > tbody > tr > td > a").each do |a|
  # Skip season links
  next if (a/"b").any?

  # Download a transcript
  season, episode = a.previous.previous.inner_text.strip.split(".")
  title = a.inner_text.strip
  url = a["href"]

  puts "Download %s.%s %s" % [season, episode, title]

  page = URI.open(url) { |f| Nokogiri::HTML(f) }  
  transcript = Nokogiri::HTML(page.css("div.entryText.s2-entrytext").inner_html.gsub!(/<br\s*\/?>/, "\n").strip).text.strip

  # Write the transcript to a file
  File.open("#{season}-#{episode}.txt", "w") do |f|
    f.write("S#{season}E#{episode}: #{title}\n")
    f.write(transcript)
  end
end

Note that this was based on https://gist.github.com/eungju/767906 and "updated" using ChatGPT. I had about a total of 10 minutes experience with Ruby code.